JPH06121393A - Loudspeaking method and loudspeaker - Google Patents

Abstract

PURPOSE:To obtain an excellent articulation with a comparatively small loudspeaking sound level even when a noise level is high. CONSTITUTION:An equalizer 38 inputs an announcement sound 40 outputted from a microphone 32 or a reproduction device 34 to perform equalizing on a DOWN side on a broad band with its spectrum peck frequency (nearly 250Hz) as a center. A sound volume adjustment means 42 adjustment signal 40 with sound volume adjusted is amplified by a announcement signal 40 with sound volume adjusted is amplified by a power amplifier 44 and the sound radiates from speakers 48 arranged in a space 46 to be loudspoken. A sound volume adjustment control means 60 decreases the sound volume when a noise level is small and increases the sound volume when the noise level is large, decreases relatively the sound volume when a noise frequency characteristic is of a lower or medium frequency band type and increases relatively the sound volume when the noise frequency characteristic is of a higher frequency band type.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a loudspeaker and a loudspeaker used for loudspeaking such as announcements in a noisy environment, with improved sound clarity (easiness of hearing including clarity). is there.

[0002]

2. Description of the Related Art In order to accurately transmit predetermined voice information in a noisy environment, it is necessary to set an optimum loud sound level.

Conventionally, a noise level fluctuation is detected, and the level of the loud sound is increased or decreased according to the noise level so that the loud sound is always produced at a constant S / N regardless of the noise level.

[0004]

In the conventional method for increasing the level of the loud sound according to the noise level, the main band of the loud sound is masked depending on the type of noise, and it is difficult to hear even if the volume is increased. The phenomenon was happening.
In addition, since the sound is amplified so that the S / N is constant with respect to the noise, it is easy to cause an adverse effect that the ambient noise is increased by the sound of the loud sound.

The present invention solves the above-mentioned problems in the prior art, and provides a loudspeaking method and a loudspeaker capable of obtaining good clarity at a relatively low loudness level even when the noise level is high. Is what you are trying to do.

[0006]

The loudspeaking method of the present invention comprises:
The level of the frequency band near the spectrum peak frequency of the signal to be amplified is relatively lowered so that the signal is amplified.

Further, the loudspeaker according to the present invention inputs a signal to be loudened and relatively lowers the level of the frequency band near the spectrum peak frequency, and the input side of the frequency characteristic adjusting means. Alternatively, at least one side of the output side, the volume adjusting means for adjusting the volume of the whole signal to be amplified, the noise level detecting means for detecting the noise level of the place where the signal to be amplified is to be amplified, and the noise level A noise frequency characteristic detecting means for detecting a frequency characteristic, and a volume is reduced when the level of the detected noise is low, a volume is increased when the level of the noise is high, and a frequency characteristic of the detected noise is low. The sound volume adjusting means is controlled so that the sound volume is relatively reduced when the sound frequency is close to the range, and is relatively increased when the frequency characteristic of noise is close to the high range. Volume adjustment control means, power amplification means for power-amplifying the frequency-characteristic-adjusted and volume-adjusted signal to be amplified, and sound-enhancing means for expanding the power-amplified signal to be amplified in the space to be amplified. And is provided.

[0008]

According to the experiments conducted by the inventors, it was found that the clarity of the loud sound can be improved by relatively lowering the level of the frequency band near the spectrum peak frequency of the signal to be louder. According to the loudspeaking method of the present invention, by utilizing the above phenomenon, the level of the frequency band in the vicinity of the spectrum peak frequency of the signal to be loudly loudened is relatively lowered so that the loudness of the loud sound is improved. Can be made. Therefore, even if the noise level is high, a relatively low level of the loud sound is sufficient, and the adverse effect on the surrounding environment can be suppressed.

Further, according to the loudspeaker of the present invention, the level of noise and the frequency characteristic are detected by paying attention to the property that the clarity of the loud sound changes according to not only the noise level but also the frequency characteristic of the noise. Since the loudness level is automatically controlled accordingly, it is possible to always obtain appropriate intelligibility regardless of the type and magnitude of noise.

[0010]

An embodiment of the present invention will be described below. First, a hearing test conducted by the inventors for making the present invention will be described. The audibility test was carried out in a simple anechoic chamber using the device shown in FIG. In the simple anechoic chamber 10, a loudspeaker 12 for loudspeaker is arranged toward the subject 14, and a speaker 16 for noise is arranged opposite to the subject toward the wall. The sound to be amplified is reproduced from the tape recorder 18, amplified by the power amplifier 22 via the digital equalizer 20, and emitted from the speaker 12. As for noise, pink noise generated by the noise generator 24 is amplified by the power amplifier 28 via the digital equalizer 26, and radiated from the speaker 16. An attenuator 30 is arranged in the hand of the subject 14, and the level of the loud sound is adjusted by operating the attenuator 30.
Can be adjusted freely.

In the hearing test, when the frequency characteristic of the loud sound is set by the digital equalizer 20 and the noise level is variously changed, the subject 14 himself adjusts the volume level of the loud sound with the attenuator 30 to obtain the optimum level. It was carried out by an adjusting method for measuring the upper and lower limits that can be felt.

FIG. 3 shows a typical frequency characteristic of a noise environment in which it is considered necessary to control the loud sound. Based on their characteristics, these are roughly classified into two types: low range type (Low (station platform, train running sound), etc.) and mid range type (Mid (station concourse, crowd in exhibition hall, etc.). In the hearing test,
High range type: High is added to these, and stationary noise in which pink noise generated from the noise generator 24 has the characteristics shown in FIG. 4 by the equalizer 26 is used as a sound source.

As the loud sound reproduced from the tape recorder 18, continuous voices (news) of male and female having the frequency characteristics shown in FIG. 5 are used, and the equalizer 26 is used to center the peak frequency of 250 Hz. In the wide band shown in FIG. 6, equalizing of −10, −5, 0, +5, +10 dB was performed, and the hearing test was repeated.

The optimum level of the loud sound is expected to have a certain range including individual differences. Therefore, in order to clarify this width, the upper limit value and the lower limit value of the optimum loud sound level were obtained by the adjusting method using the index shown in FIG. The subjects were three adult males with normal hearing.

According to the above-mentioned hearing test, the optimum range of the loud sound level with respect to the noise level is roughly as shown in FIG. 8. Here, it is necessary to ensure the clarity with the minimum necessary volume. Since this is the purpose, examine the test results by focusing only on the lower limit of the optimum level.

FIG. 9 shows the measurement result of the lower limit value of the optimum level of the loud sound when the frequency characteristic of the noise is set to the low tone range Low. According to this, when the loud sound is equalized to the DOWN (-dB) side, it is recognized that the same intelligibility is ensured even when the volume is about 3 dB lower than the reference value (0 dB). On the other hand, in equalizing on the UP (+ dB) side, it is necessary to raise the volume to obtain the same clarity, which is contrary to the purpose of ensuring the clarity with the minimum necessary volume. These tend to be the same for both male and female voices.

Next, FIG. 10 shows the test results in the male voice expressed by the level difference from the reference value for each frequency characteristic of noise.
It shows in (a)-(c). According to this, the loud sound is DOW
Comparing the improvement effects when equalizing to the N side, the noise is about 3 dB for Low and Mid,
Noise: About 1 dB at High. This indicates that the equalization of the loud sound to the DOWN side is more effective for the low-pitched sound and the mid-range noise, and if it is applied to the noise environment as shown in FIG. It suggests that the sound can be minimized.

As described above, as a result of examining the possibility of securing the clarity by the equalizing of the loud sound, as a result of the hearing test, the equalizing to the DOWN side is performed in a wide band centering on the peak frequency of the loud sound. As a result, it became clear that the volume can be suppressed while maintaining the clarity. It was also found that this method is more effective for low-pitched and mid-range noises and can be particularly effectively used for loudspeaking in such a noise environment.

Although the reason why the above test results are obtained is not always clear, there are many vowel components around the peak of the human voice, and there are consonant repellency and plosive sounds in the high range. It is speculated that lowering the vowel level enhances the consonant more than one reason for improving the clarity. Further, if the vowel component itself in the loud sound is also a cause of noise, it is considered that lowering the vowel level in the loud sound lowers the overall noise level.

Next, FIG. 1 shows an embodiment of the loudspeaker of the present invention utilizing the above test results. A raw announcement sound by a human voice is input from the microphone 32. An announcement sound by a human voice is reproduced from the reproducing device 34 such as a tape recorder. The announcement signal input from the microphone 32 is amplified by the microphone amplifier. Equalizer 38 is microphone 32
Alternatively, an announcement sound (a signal to be amplified) 40 output from the playback device 34 is input and flattened or equalized to the DOWN side with the characteristic shown in FIG. 6 in a wide band centered on the spectrum peak frequency (about 250 Hz). To do. The volume adjusting means 42 adjusts the overall volume of the signal 40 to be amplified. Volume-adjusted announcement signal 40
Is a space 46 to be amplified by the power amplifier 44 and to be loud.
It is radiated from the speaker 48 arranged at.

A microphone 50 for picking up noise is arranged in the space 46 to be amplified. The output of the microphone 50 is amplified by the amplifier 51 and input to the noise level detecting means 52 and the noise frequency characteristic detecting means 54. The noise level detecting means 52 detects the noise level in the space 46 and outputs it as a value averaged at regular intervals. Since the noise level detection detects the overall level of noise, no special filter is required, but when considering control by A characteristic level detection as a simple system, it is possible to incorporate an A characteristic filter, etc. Is. The frequency characteristic detection means 54 determines whether the frequency characteristics of the collected noise are low-frequency type (Low) or middle-frequency type (Mid).
It is detected whether it is a high frequency type (High).

The equalizing control means 58 formed in the CPU 56 sets the characteristic of the equalizer 38 in accordance with the level frequency characteristic of noise, if necessary. DOWN
It is considered effective to set the volume as large as possible within the range where the naturalness of the announcement is maintained, but if it is set too large, unnaturalness will be emphasized and it will be difficult to hear, so in consideration of the sound quality, DOWN Set the amount. According to the above test results, it may be set to -5 dBDOWN or -10 dBDOWN. In addition, when the noise level is low and the loud sound level is low, it is not necessary to worry about the influence on the surrounding environment, and therefore the sound quality (ease of hearing) is emphasized and flat (0 dB) characteristics may be used. Further, this equalizing control can be performed manually instead of automatically.

The volume adjusting control means 60 controls the volume adjusting means 42 according to the noise level and the frequency characteristic to automatically adjust the loud sound level as shown in FIGS. 9 and 10. That is, when the noise level is low, the volume is decreased, when the noise level is high, the volume is increased, and when the noise frequency characteristic is low-range or mid-range, the volume is relatively low (-3 dB. When the frequency characteristic of noise is high frequency type, the volume is relatively increased (-1 dB). As a result, a clear loud sound can always be obtained with the minimum necessary volume regardless of the noise level and frequency characteristics, and the adverse effect on the surrounding environment can be suppressed. The volume correction volume 62 can give a volume correction amount in advance according to the surrounding environment, the installation position of the speaker, and the like.

[0024]

Other Embodiments In the above embodiment, the equalizing center frequency of the equalizer 38 is fixed, but the signal 4 to be amplified is
If the spectrum peak frequency of 0 fluctuates, the equalizing center frequency can also fluctuate accordingly. FIG. 11 shows an embodiment thereof, in which the peak frequency detecting means 70 detects the spectrum peak frequency of the signal 40 to be amplified and controls the equalizing center frequency of the equalizer 38 to match the frequency. As a result, control suitable for loud sound with various frequency characteristics is realized.

Further, the frequency characteristics and noise level of the loud sound,
If the frequency characteristic is constant, it can be simplified as shown in FIG. That is, the signal 40 to be amplified which is input from the microphone 32 is amplified by the amplifier 36, is given a fixed characteristic (for example, the characteristic of −10 dB in FIG. 6) by the equalizer 38, is amplified by the power amplifier 44, and is a speaker. Loud at 48.

[0026]

As described above, according to the loudspeaking method of the present invention, since the level of the frequency band near the spectrum peak frequency of the signal to be loudened is relatively lowered and the loudspeaking sound is produced. The clarity of can be improved. Therefore, even if the noise level is high, a relatively low level of the loud sound is sufficient, and the adverse effect on the surrounding environment can be suppressed.

According to the loudspeaker of the present invention, the noise level and the frequency characteristic are detected by paying attention to the property that the clarity of the loud sound changes according to not only the noise level but also the frequency characteristic of the noise. Since the loudness level is automatically controlled accordingly, it is possible to always obtain appropriate intelligibility regardless of the type and magnitude of noise.

[Brief description of drawings]

FIG. 1 is a block diagram showing an embodiment of the present invention.

FIG. 2 is a block diagram showing a hearing test apparatus.

FIG. 3 is a diagram showing typical frequency characteristics of a noise environment in which it is considered necessary to control a loud sound.

FIG. 4 is a characteristic diagram of noise used in a hearing test.

FIG. 5 is a diagram showing frequency characteristics of continuous voices (news) of men and women.

FIG. 6 is an equalizing characteristic diagram.

FIG. 7 is a diagram showing an index of an optimum level of a loud sound.

FIG. 8 is a diagram showing an optimum range of a loud sound level with respect to a noise level measured by a hearing test.

FIG. 9 is a diagram showing the measurement result of the lower limit value of the optimum level of the loud sound when the frequency characteristic of noise is set to the low sound range Low.

FIG. 10 is a diagram showing a test result in a male voice as a level difference from a reference value for each frequency characteristic of noise.

FIG. 11 is a block diagram showing another embodiment of the present invention.

FIG. 12 is a block diagram showing still another embodiment of the present invention.

[Explanation of symbols]

 38 Equalizer (frequency characteristic adjusting means) 40 Signal to be amplified 42 Volume adjusting means 44 Power amplifying means 48 Speaker (speaking means) 52 Noise level detecting means 54 Noise frequency characteristic detecting means

Claims (2)

[Claims] 1. A loudspeaking method which relatively lowers the level of a frequency band in the vicinity of a spectrum peak frequency of a signal to be loudspeaked. 2. A frequency characteristic adjusting means for inputting a signal to be amplified and relatively lowering the level of a frequency band near the spectrum peak frequency, and at least one of the input side and the output side of the frequency characteristic adjusting means. A volume adjusting means for adjusting the volume of the entire signal to be amplified, a noise level detecting means for detecting a noise level of a place where the signal to be amplified is to be amplified, and a noise frequency for detecting a frequency characteristic of the noise. A characteristic detecting means, and lowers the volume when the level of the detected noise is low, increases the volume when the level of the noise is high, and decreases the volume when the frequency characteristic of the detected noise is near the low range. Volume adjustment control means for controlling the volume adjustment means so that the volume is relatively increased and the volume is relatively increased when the frequency characteristic of noise is in the high frequency range, A loudspeaker comprising power amplification means for power-amplifying the signal to be amplified, the frequency characteristic of which has been adjusted and the volume adjusted, and a loudspeaker means for amplifying the power-amplified signal to be amplified, in the space to be amplified. apparatus.